Add option to use utf-16 positions instead of utf-8

We use this parser for a custom linter that we ship through a VSCode extension (rustpython_parser::ast::Suite::parse_without_path). The VSCode extension api expects byte positions to the text encoded in utf-16, but the parser returns the byte positions to text encoded in utf-8. Can we add an option to use utf-16 encoded byte positions?

I'd be happy to implement this feature if you're comfortable with adding it to the parser. Do you have suggestions on how the API should look like?

从 rustpython_parser::ast::Suite::parse_without_path 开始，追踪源代码位置是如何生成和暴露的。检查现有的 parser API 和与位置相关的测试，然后确定一种用于选择 UTF-16 位置的 API 形式。完成的标准是该选项已实现，并且其位置行为已由测试覆盖。